Home > Academics > Undergraduate > Undergrad Research Opportunities > 2021-2022 SURE Research Projects in CSE

2021-2022 SURE Research Projects in CSE

This page lists summer research opportunities in CSE that are available through the SURE Program. To learn more or apply, visit: https://sure.engin.umich.edu/.

Directions

  • Please carefully consider each of the following projects, listed below, before applying to the SURE Program.
  • You must indicate your top three project choices on your SURE application, in order of preference, using the associated CSE project number.
  • Questions regarding specific projects can be directed to the listed faculty mentor. 
  • Timeline: SURE applications will be reviewed throughout the month of March and recipients will be notified sometime in late-March/early-April.

Project descriptions

CSE Project #1: Ubiquitous Health Sensing
Faculty Mentor: Alanson Sample [apsample @ umich.edu] 
Prerequisites: Experience with embedded systems, computer vision, or machine learning.
Description: Effective means of unobtrusive and continuous monitoring of one’s health could transform how we detect and treat illnesses. This project aims to create a long-range health monitoring system that can passively measure an individual’s vital signs and daily activities from a distance of up to three meters. Building off of novel sensing techniques developed in the Interactive Sensing and Computing Lab, SURE students will work with faculty and graduate student mentors to create a fully working end-to-end system, utilizing embedded systems, computer vision, and machine learning.
Expected research delivery mode: In-person

CSE Project #2: Privacy-Preserving Sensing for in-home Activity Recognition
Faculty Mentor: Alanson Sample [apsample @ umich.edu]  
Prerequisites: Experience with embedded systems, computer vision, or machine learning.
Description: Giving computers the ability to sense and understand our daily activities and routines can enable new smart home applications, context-aware computing, and transform how we detect, monitor, and treat disease and chronic illnesses. However, existing smart devices rely on remote cloud services to process voice commands, analyze video, and perform recognition tasks. Even if these smart devices record only ML features to send to the cloud, private data still leaves the home for cloud-based processing and is stored for an indeterminate amount of time. This project aims to create a new class of smart sensors and embedded devices that removes Personally Identifiable Information before sensitive data leaves the devices while maintaining downstream activity recognition applications. Examples include microphones the remove speech but leave other acoustic information for audio classification and cameras that use onboard GPUs to remove and replace images of people robustly.
Expected research delivery mode: In-person

CSE Project #3: Hybrid 60GHz Radar and Computer Vision Sensing
Faculty Mentor: Alanson Sample [apsample @ umich.edu]  
Prerequisites: Some experience in ONE of the following is preferred: Computer Vision, Machine Learning, Radar, Software Defined Radios.
Description: Computer vision offers a high-fidelity means for devices to sense and understand the world around them. However, cameras suffer from occlusion, are dependent on ambient lighting conditions, and raise privacy concerns when placed in the home or sensitive areas. In contrast, emerging 5G and mmWave radio systems offer unique sensing capabilities that allow devices to see through walls, peer into the human body, and sense user actions to enable new computer interfaces. This project will focus on creating a hybrid computer vision and 60GHz radar system that will leverage the best of both worlds. SURE students will work with faculty and graduate mentors to explore applications in the areas of indoor localization and pose estimation, indoor activity detection, and sensing health parameters such as heart rate, breath rate, and lung capacity.
Expected research delivery mode: In-person

CSE Project #4: Improving the Performance of Modern Data Center Systems
Faculty Mentor: Baris Kasikci [barisk @ umich.edu] 
Prerequisites: EECS 482 or equivalent, strong C++ programming skills
Description: Modern data-center applications suffer significant slow-down due to bottlenecks in the memory system as well as in the processors’ instruction supply. As part of this project we will develop hardware/software co-design techniques to (1) detect, (2) classify, (3) eliminate such bottlenecks. We will design new compiler techniques, operating systems support, and hardware extensions in order to build efficiency next generation data center systems. This line of work has been very fruitful, leading to multiple paper publications for SURE students over the past three years. These past SURE students have since gone on to top PhD programs in the world as well as to positions in major companies. The goal in this year’s SURE project is to similarly produce cutting edge research.
Expected research delivery mode: Hybrid

CSE Project #5: Hazel Notebooks: Building a Better Jupyter
Faculty Mentor: Cyrus Omar [comar @ umich.edu] 
Prerequisites: EECS 490 or equivalent is preferred, but not required.
Description: The popular Jupyter lab notebook environment is powerful, but it has a problem: results stored in a notebook are not reproducible, because the user can execute cells out of order. In our group, we are developing a new live functional programming environment called Hazel (hazel.org). Right now, Hazel does not support multiple program cells. This project will turn Hazel into a next-generation version of Jupyter by adding support for notebooks with multiple cells, with dependencies between them. We will solve the reproducibility problem by developing a mechanism conjectured in a recent paper in our group: fill-and-resume.
Expected research delivery mode: Too soon to say

CSE Project #6: Hazel: A Live Functional Programming Environment
Faculty Mentor: Cyrus Omar [comar @ umich.edu] 
Prerequisites: EECS 490 or equivalent is preferred, but not required.
Description: Hazel (hazel.org) is a live functional programming environment that is able to typecheck, transform and even execute incomplete programs, i.e. programs with holes. There are a number of projects available within the Hazel project for a student interested in research into programming languages.
Expected research delivery mode: Too soon to say

CSE Project #7: Computer Vision for Physical and Functional Understanding
Faculty Mentor: David Fouhey [fouhey @ umich.edu] 
Prerequisites: Good grades in EECS 442 OR EECS 445.
Description: The lab is broadly focused on building 3D representations of the world and understanding human/object interaction. Potential projects include learning about: navigating environments, object articulations, commonsense physical properties of objects, and hand grasps. Please look at:http://web.eecs.umich.edu/~fouhey/ for a sense of what projects we’ve done in the past. We will find a specific project based on mutual interest and particular abilities (e.g., stronger systems programming abilities, experience with graphics, etc.). Students looking for a longer term project continuing during the school year are strongly encouraged to apply.
Expected research delivery mode: Too soon to say

CSE Project #8: Collaborative Models for Grounded Language Processing
Faculty Mentor: Joyce Chai [chaijy @ umich.edu] 
Prerequisites: EECS 492, EECS 445, proficiency in python programming, knowledge and background in computer vision and natural language processing.
Description: In human-robot communication, although humans and physical agents are co-present in a shared environment, they have mismatched abilities in perceiving the shared environment and reasoning about joint tasks. Their representations are significantly misaligned, which makes language communication difficult. To address this problem, this project will investigate language use in situated communication for collaborative tasks. It will develop computational models that integrate language and dialogue processing to allow humans and agents to mediate disparities in their representations and enable collaborative language grounding.
Expected research delivery mode: Too soon to say.

CSE Project #9: Hierarchical Task Learning from Language Instructions
Faculty Mentor: Joyce Chai [chaijy @ umich.edu] 
Prerequisites: EECS 492, EECS 445, proficiency in python programming, knowledge and background in computer vision and natural language processing.
Description: Language instructions play an important role in human learning and task acquisition. To enable similar abilities in AI agents, there has been an increasing amount of work on task learning from natural language (NL) instructions and expert demonstrations. Despite recent progress, task learning from natural language instructions remains an extremely challenging problem. Most approaches directly map instructions to primitive actions in an end-to-end architecture, which suffers from generalizability and transparency. To address these issues, this project will systematically investigate hierarchical task learning that ties high-level abstract (sub)goals with low-level grounded primitive actions from language instructions and embodied experience with the physical world.
Expected research delivery mode: Too soon to say.

CSE Project #10: Collaborative Intelligence between Mobile and Cloud
Faculty Mentor: Lingjia Tang [lingjia @ umich.edu] 
Prerequisites: EECS 482.
Description: The goal of the project is to design generalized and effective collaborative intelligence approaches that partition computation between mobile devices and cloud for intelligent applications such as virtual assistants, smart home and autonomous vehicle. We want to target dynamic, complex edge/grid/cloud environments to support the state-of-the-art large-scale data-intensive, compute-intensive intelligent applications and achieve low latency, high accuracy and high energy efficiency.
Expected research delivery mode: Hybrid.

CSE Project #11: Natural Language Processing for Understanding Media Bias and Fake News
Faculty Mentor: Lu Wang [wangluxy @ umich.edu] 
Prerequisites: EECS 445 (Machine Learning), probability and statistics, experience with natural language processing problems, proficient in Python.
Description: News media play a vast role not just in supplying information, but in selecting, crafting, and biasing that information to achieve both nonpartisan and partisan goals. We aim to automate media bias detection from news articles, and quantify and further highlight biased content in order to promote the transparency of news production as well as enhance readers’ awareness of media bias. This project will explore and design natural language processing and machine learning algorithms to detect media bias. Specifically, we will work on developing information extraction systems, e.g., important entities and narrative structure will be extracted automatically from news articles. The developed tools will also be used for understanding fake news.
Expected research delivery mode: Hybrid.

CSE Project #12: Does Wealth Matter? Learning Generative Models with Prediction Markets
Faculty Mentor: Mithun Chakraborty + Sindhu Kutty [dcsmc @ umich.edu] 
Prerequisites: EECS 445 and STATS 412 (or equivalents) preferred.
Description: As recent events have highlighted, polling can be messy, misleading and prone to misinterpretation. Markets have the advantage over polls in having built-in financial incentives and timely responses, and have been empirically observed to outperform alternative forecasting tools such as polls. However, when traders have varying degrees of wealth, are markets egalitarian? Moreover, how precise are they and what factors impact their precision? We will answer these questions in the context of Prediction Markets by tying market prices to learning a generative model of the outcome space. We will also explore other connections between convergence in Machine Learning algorithms (especially Bayesian processes) and equilibria in these markets.

Prediction markets (e.g. Iowa Electronic Markets, PredictIt, etc.) are a type of financial market the purpose of which is to elicit the personal beliefs of traders about a future uncertain event and aggregate these beliefs into the market price. In this project, students will implement and execute a set of experiments on the interaction of a new prediction market design with simulated trading agents having diverse risk attitudes and help address the above research questions in different environments in a systematic manner. An understanding of connections to Machine Learning algorithms would be illustrative for gauging the accuracy, and hence reliability, of Prediction Markets and can, in turn, inform innovations in their design. The learning outcome for students will be hands-on experience in interdisciplinary research with connections to Machine Learning and Computational Economics.
Expected research delivery mode: Too soon to say.

CSE Project #13: Understanding Gig Workers’ Experiences with AI Algorithms
Faculty Mentor: Nikola Banovic [nbanovic @ umich.edu] 
Prerequisites: Strong interest in qualitative research methods such as interviewing people, observing activities, etc.
Description: AI has started to transform the nature of work in many sectors of the economy. One of the most tangible transformations has been in the on-demand economy, for services such as grocery delivery, ride-hailing, and other last-mile services, where its advances have allowed a shift towards greater efficiency, through the use of AI-mediated platforms. On-demand work, with its promises of flexibility, independence and entrepreneurship is also an attractive option for individuals seeking a low-barrier entry into employment and economic opportunities. However, several recent debates around the employment status of workers with services such as Uber, Lyft and Instacart have shined a light on the adversarial relationships between workers and platforms, and the negative effects of opaque algorithms on workers’ well-being. In this project, we seek to work with the community of Instacart shoppers to uncover algorithmic interaction patterns that may be harmful to them, and to develop models of algorithmic accountability. Our goal is to inform the design of platforms that enhance worker well-being and their access to economic opportunities.
Expected research delivery mode: Too soon to say.

CSE Project #14: Fundamental Understanding of Deep Learning
Faculty Mentor: Wei Hu [vvh @ umich.edu] 
Prerequisites: EECS 445 (machine learning); solid background in linear algebra, calculus, probability and statistics; strong programming skills; familiarity with a deep learning library (e.g. PyTorch, JAX, TensorFlow) is desirable but not required.
Description: Despite the phenomenal successes of deep learning methods in various AI application domains, their underlying working mechanisms remain poorly understood. This project aims to obtain a deeper fundamental understanding of the properties of deep learning methods, and potentially to inform new practical advancements. Possible directions include but are not limited to the studies of training dynamics, implicit biases, representations in neural networks, pre-training and transfer learning, robustness, fairness, out-of-distribution generalization, and architectural components. The project can be in the form of either theoretical analysis, or empirical investigation, or both. Students who intend to continue beyond summer and perform long-term research are welcome.
Expected research delivery mode: Too soon to say.

CSE Project #15: Web Automation using Program Synthesis (Front-end)
Faculty Mentor: Xinyu Wang [xwangsd @ umich.edu] 
Prerequisites: EECS 485 or familiarity with HTML/DOM/Javascript/Typescript.
Description: Many computer end-users often need to perform tasks that involve the web, such as filling online forms and scraping data, which are repetitive and tedious in nature. On the other hand, there are existing tools and languages, such as Selenium, that can be used to automate these tasks. However, writing automation scripts is far beyond the capability of end-users who have very little programming background. In this project, we aim to help users automate web-related programming tasks using an AI technique called program synthesis. We already built an initial prototype for this project, but we’re looking to significantly expand the project. We’re looking for a few students to work on the front-end development which involves designing and implementing user interfaces.
Expected research delivery mode: Too soon to say.

CSE Project #16: Web Automation using Program Synthesis (Back-end)
Faculty Mentor: Xinyu Wang [xwangsd @ umich.edu] 
Prerequisites: EECS 203 and 280/281, and/or EECS 490/481. Experience with Rust is a plus.
Description: Many computer end-users often need to perform tasks that involve the web, such as filling online forms and scraping data, which are repetitive and tedious in nature. On the other hand, there are existing tools and languages, such as Selenium, that can be used to automate these tasks. However, writing automation scripts is far beyond the capability of end-users who have very little programming background. In this project, we aim to help users automate web-related programming tasks using an AI technique called program synthesis. We already built an initial prototype for this project, but we’re looking to significantly expand the project. We’re looking for a few students to work on the back-end development, which involves designing and implementing synthesis algorithms.
Expected research delivery mode: Too soon to say.

CSE Project #17: Superoptimizing SQL Queries
Faculty Mentor: Xinyu Wang [xwangsd @ umich.edu] 
Prerequisites: EECS 484 (or familiarity with SQL), or EECS 481 (software engineering), or EECS 483 (compilers).
Description: SQL queries, if written poorly, are slow on large databases, even using state-of-the-art query optimizers. This project aims to develop a super optimizer for SQL queries, which is able to maximally boost the performance of a poorly written query.
Expected research delivery mode: Too soon to say.

CSE Project #18: Automatically Generating TensorFlow Programs
Faculty Mentor: Xinyu Wang [xwangsd @ umich.edu] 
Prerequisites: EECS 280/281. Experience with TensorFlow is a plus.
Description: As machine learning becomes increasingly popular, more and more developers start to use frameworks such as TensorFlow. However, it’s not at all easy to program in these frameworks, partly because such programs involve transforming high-dimensional arrays (i.e., tensors and matrices). In this project, we aim to develop a tool that significantly eases the development of tensor manipulation programs.
Expected research delivery mode: Too soon to say.

CSE Project #19: Human-AI Collaborative Methods for Instructional Design
Faculty Mentor: Xu Wang [xwanghci @ umich.edu] 
Prerequisites: Proficiency in web development is required (e.g., have taken 485 and 493), skills on human-centered design is preferred (493 or SI classes on need assessment, e.g., conducting interviews, creating prototypes).
Description: In this project, we will be developing tools to support instructors in their instructional design process. We will develop web-based platforms that provide data-driven intelligent support powered by natural language processing techniques to help instructors create assessments, design learning materials (e.g. lecture slides), develop grading rubrics, etc. Students participating in this project will get exposure to a full pipeline of HCI project and learn and exercise skills on human-centered design and development and applied machine learning.
Expected research delivery mode: In-person.

CSE Project #20: Developing Mixed-reality Tutoring System to Support Medical Education
Faculty Mentor: Xu Wang [xwanghci @ umich.edu] 
Prerequisites: Strong software engineering skills preferred, 485 required. Familiarity with Unity a plus (or willingness to learn Unity).
Description: Medical procedures such as heart resuscitation are often complex and require intensive decision making in highly stressful contexts. Clinicians can be under prepared for these procedures if they don’t frequently encounter patients that require such treatment. In this project, we will develop Augmented reality systems that support clinicians in learning and performing such complex medical procedures. Students participating in this project will get full exposure to the pipeline of an HCI project and learn and exercise human-centered design and development, development of augmented reality software, and applied machine learning techniques.
Expected research delivery mode: In-person.